skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Application of a new hybrid particle swarm optimization-mixed kernels function-based support vector machine model for reservoir porosity prediction: A case study in Jacksonburg-Stringtown oil field, West Virginia, USA

Abstract

Porosity is a fundamental property that characterizes the storage capability of fluid and gas-bearing formations in a reservoir. An accurate porosity value can be measured from core samples in the laboratory; however, core analysis is expensive and time consuming. Well-log data can be used to calculate porosity, but the availability of log suites is often limited in mature fields. Therefore, robust porosity prediction requires integration of core-measured porosity with available well-log suites to control for changes in lithology and fluid content. A support vector machine (SVM) model with mixed kernel function (MKF) is used to construct the relationship between limited conventional well-log suites and sparse core data. Porosity is the desired output, and two conventional well-log responses (gamma ray [GR] and bulk density) and three well-log-derived parameters (the slope of GR, the slope of density, and [Formula: see text]) are input parameters. A global stochastic searching algorithm, particle swarm optimization (PSO), is applied to improve the efficiency of locating the appropriate values of five control parameters in MKF-SVM model. The results of SVM with different traditional kernel functions were compared, and the MKF-SVM model provided an improvement over the traditional SVM model. To confirm the advantage of the hybrid PSO-MKF-SVMmore » model, the results from three models: (1) radial basis function (RBF)-based least-squares SVM, (2) multilayer perceptron artificial neural network (ANN), and (3) RBF ANN, are compared with the result of the hybrid PSO-MKF-SVM model. The results indicate that the hybrid PSO-MKF-SVM model improves porosity prediction with the highest correlation coefficient ([Formula: see text] of 0.9560), the highest coefficient of determination ([Formula: see text] of 0.9140), the lowest root-mean-square error (1.6505), average absolute error value (1.4050), and maximum absolute error (2.717).« less

Authors:
 [1];  [2]
  1. West Virginia University, Department of Geology and Geography, West Virginia, Morgantown 26506, USA and The University of Texas at Austin, Bureau of Economic Geology, Texas, Austin 78713, USA.(corresponding author).
  2. West Virginia University, Department of Geology and Geography, West Virginia, Morgantown 26506, USA..
Publication Date:
Research Org.:
West Virginia Univ., Morgantown, WV (United States)
Sponsoring Org.:
USDOE Office of Policy and International Affairs (PO)
OSTI Identifier:
1613809
DOE Contract Number:  
PI0000017
Resource Type:
Journal Article
Journal Name:
Interpretation
Additional Journal Information:
Journal Volume: 7; Journal Issue: 1; Journal ID: ISSN 2324-8858
Publisher:
Society of Exploration Geophysicists
Country of Publication:
United States
Language:
English
Subject:
Geochemistry & Geophysics

Citation Formats

Zhong, Zhi, and Carr, Timothy R. Application of a new hybrid particle swarm optimization-mixed kernels function-based support vector machine model for reservoir porosity prediction: A case study in Jacksonburg-Stringtown oil field, West Virginia, USA. United States: N. p., 2019. Web. doi:10.1190/int-2018-0093.1.
Zhong, Zhi, & Carr, Timothy R. Application of a new hybrid particle swarm optimization-mixed kernels function-based support vector machine model for reservoir porosity prediction: A case study in Jacksonburg-Stringtown oil field, West Virginia, USA. United States. https://doi.org/10.1190/int-2018-0093.1
Zhong, Zhi, and Carr, Timothy R. 2019. "Application of a new hybrid particle swarm optimization-mixed kernels function-based support vector machine model for reservoir porosity prediction: A case study in Jacksonburg-Stringtown oil field, West Virginia, USA". United States. https://doi.org/10.1190/int-2018-0093.1.
@article{osti_1613809,
title = {Application of a new hybrid particle swarm optimization-mixed kernels function-based support vector machine model for reservoir porosity prediction: A case study in Jacksonburg-Stringtown oil field, West Virginia, USA},
author = {Zhong, Zhi and Carr, Timothy R.},
abstractNote = {Porosity is a fundamental property that characterizes the storage capability of fluid and gas-bearing formations in a reservoir. An accurate porosity value can be measured from core samples in the laboratory; however, core analysis is expensive and time consuming. Well-log data can be used to calculate porosity, but the availability of log suites is often limited in mature fields. Therefore, robust porosity prediction requires integration of core-measured porosity with available well-log suites to control for changes in lithology and fluid content. A support vector machine (SVM) model with mixed kernel function (MKF) is used to construct the relationship between limited conventional well-log suites and sparse core data. Porosity is the desired output, and two conventional well-log responses (gamma ray [GR] and bulk density) and three well-log-derived parameters (the slope of GR, the slope of density, and [Formula: see text]) are input parameters. A global stochastic searching algorithm, particle swarm optimization (PSO), is applied to improve the efficiency of locating the appropriate values of five control parameters in MKF-SVM model. The results of SVM with different traditional kernel functions were compared, and the MKF-SVM model provided an improvement over the traditional SVM model. To confirm the advantage of the hybrid PSO-MKF-SVM model, the results from three models: (1) radial basis function (RBF)-based least-squares SVM, (2) multilayer perceptron artificial neural network (ANN), and (3) RBF ANN, are compared with the result of the hybrid PSO-MKF-SVM model. The results indicate that the hybrid PSO-MKF-SVM model improves porosity prediction with the highest correlation coefficient ([Formula: see text] of 0.9560), the highest coefficient of determination ([Formula: see text] of 0.9140), the lowest root-mean-square error (1.6505), average absolute error value (1.4050), and maximum absolute error (2.717).},
doi = {10.1190/int-2018-0093.1},
url = {https://www.osti.gov/biblio/1613809}, journal = {Interpretation},
issn = {2324-8858},
number = 1,
volume = 7,
place = {United States},
year = {2019},
month = {2}
}

Works referenced in this record:

Support vector regression for porosity prediction in a heterogeneous reservoir: A comparative study
journal, December 2010


Support-Vector Regression for Permeability Prediction in a Heterogeneous Reservoir: A Comparative Study
journal, June 2010


Fuzzy classifier based support vector regression framework for Poisson ratio determination
journal, September 2013


Robust linear programming discrimination of two linearly inseparable sets
journal, January 1992


Neural Networks and the Bias/Variance Dilemma
journal, January 1992


Evaluation of Acid Fracturing Treatments in Shale Formation
journal, September 2017


Porosity and permeability prediction from wireline logs using artificial neural networks: a North Sea case study
journal, July 2001


Reservoir oil viscosity determination using a rigorous approach
journal, January 2014


Estimating the initial pressure, permeability and skin factor of oil reservoirs using artificial neural networks
journal, January 2006


Reservoir properties determination using fuzzy logic and neural networks from well data in offshore Korea
journal, December 2005


What is a support vector machine?
journal, December 2006


Particle swarm optimization: An overview
journal, August 2007


Reservoir Characterization by a Combination of Fuzzy Logic and Genetic Algorithm
journal, February 2014


Influence of gravel on the propagation pattern of hydraulic fracture in the glutenite reservoir
journal, June 2018


A new model to evaluate two leak points in a gas pipeline
journal, October 2017


Investigation into the performance of oil and gas projects
journal, February 2017


Design of neural networks using genetic algorithm for the permeability estimation of the reservoir
journal, October 2007


Application of hybrid neural particle swarm optimization algorithm for prediction of MMP
journal, January 2014


What is noise?
journal, July 1998


Comparing support vector machines with Gaussian kernels to radial basis function classifiers
journal, January 1997


A tutorial on support vector regression
journal, August 2004


Recurrent least squares support vector machines
journal, July 2000

  • Suykens, J. A. K.; Vandewalle, J.
  • IEEE Transactions on Circuits and Systems I: Fundamental Theory and Applications, Vol. 47, Issue 7
  • https://doi.org/10.1109/81.855471

Fault diagnostics based on particle swarm optimisation and support vector machines
journal, May 2007


Artificial neural networks in bankruptcy prediction: General framework and cross-validation analysis
journal, July 1999